← Back to Skills Marketplace
TuShare Stock Skill
by
Zexun Chen
· GitHub ↗
· v1.0.3
· MIT-0
9063
Downloads
8
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install tushare-stock-skill
Description
面向中国 A 股的 Tushare 专用技能，提供股票数据获取、个股分析与交易观察能力。
Usage Guidance
Install this only in a virtual environment, provide a Tushare token through TUSHARE_TOKEN when possible, and avoid using TUSHARE_STOCK_ENV_FILE unless it points to a minimal file containing only that token. Be aware that direct fetches can reach some adjacent non-A-share or cross-market Tushare catalog entries, and consider setting TUSHARE_STOCK_CACHE_DIR to a private per-user cache directory.
Capability Assessment
Credentials
The required TUSHARE_TOKEN, optional TUSHARE_POINTS, optional TUSHARE_STOCK_ENV_FILE, optional TUSHARE_STOCK_CACHE_DIR, and network access to Tushare are coherent for a Tushare integration. The env-file path can point anywhere if explicitly set, so users should only set it to a minimal token-only file.
Install Mechanism
Installation is a standard pip requirements flow, but requirements.txt leaves pandas, tushare, requests, and beautifulsoup4 unpinned, so installs are not reproducible and should be done in an isolated environment.
Instruction Scope
Runtime instructions are ordinary usage guidance for commands, output language, and clarification behavior. I found no instruction hijacking, hidden role changes, memory poisoning, or attempts to override the user's agent policy.
Persistence & Privilege
The skill writes a public stock-basic cache under /tmp/tushare_stock_skill by default and build_catalog.py can refresh local reference files. It does not install services, scheduled tasks, shell hooks, backdoors, or persistent token caches.
Purpose & Capability
The main script implements Tushare data fetching, natural-language endpoint routing, stock-code resolution, and stock analysis/trading-observation logic matching the stated purpose. The bundled catalog also includes adjacent Stock Connect, Hong Kong, ETF-capable, and generic pro_bar entries, which is broader than a strictly A-share-only framing but visible in the references and partly disclosed in the skill text.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tushare-stock-skill - After installation, invoke the skill by name or use
/tushare-stock-skill - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
- Updated skill name and description to focus clearly on China A-share stock research and analysis, with all user-facing text in Simplified Chinese.
- Revised documentation to clarify the core scenarios, values, usage boundaries, and exclude non-A-share market workflows.
- Streamlined examples, command references, and explanation style for more concise, practical guidance.
- No functional or code changes; this update is a full documentation and localization improvement.
v1.0.2
- Update metadata to clarify required and optional environment variables and Python dependencies
- Add explicit information on required and optional Python packages and dependent network domains
- Add a new "Installation" section with pip install instructions and environment variable usage
- Clarify boundaries by grouping environment variable descriptions and security details
- No core functionality or command changes; documentation and metadata improvements only
v1.0.1
- Major refactor: simplified project root, removed legacy folders and scripts.
- README and SKILL.md fully rewritten in concise English.
- Usage, entry points, and environment variable requirements are now clearly documented.
- Project metadata and homepage added for better discoverability.
- Chinese user output and Tushare use cases remain central, but developer docs are streamlined and focused.
v1.0.0
- 首次发布，支持全套 Tushare A 股股票数据与个股分析，包括行情、估值、财务、资金流、龙虎榜等多类接口。
- 根据自然语言智能匹配最优接口，自动识别股票与参数，可辨识查询、分析和交易观察等多类意图。
- 面向分析与交易观察请求，内置快档与深档机制，自动决定是否深扫龙虎榜和机构席位。
- 查询接口默认返回原始结构化数据，分析请求输出简要结论、核心指标及风险提示。
- 明确处理权限与积分受限接口，并采用环境变量配置 Token，避免本地 token 文件写入。
- 所有说明及输出均为简体中文，接口目录与指标体系支持扩展与自定义。
Metadata
Frequently Asked Questions
What is TuShare Stock Skill?
面向中国 A 股的 Tushare 专用技能，提供股票数据获取、个股分析与交易观察能力。 It is an AI Agent Skill for Claude Code / OpenClaw, with 9063 downloads so far.
How do I install TuShare Stock Skill?
Run "/install tushare-stock-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is TuShare Stock Skill free?
Yes, TuShare Stock Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does TuShare Stock Skill support?
TuShare Stock Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created TuShare Stock Skill?
It is built and maintained by Zexun Chen (@magica-chen); the current version is v1.0.3.
More Skills